Application of MAGNet Tools to Gene Regulatory Networks in Cancer Principal Investigator: Sander, Chris Project Summary / Abstract We propose to apply computational algorithms from the Columbia University National Center for Multiscale Analysis of Genomic and Cellular Networks (MAGNet) to study gene regulatory networks in human cancer. The project will be led by the Computational Biology Center at Memorial Sloan Kettering Cancer Center (MSKCC) and will build an intensive collaboration between computer scientists and clinical researchers. The aims of this project are to: Make discoveries about gene regulation in soft tissue sarcoma and prostate cancer. Strengthen the National Center's computational tools for the analysis of gene regulation and biological pathways, and Develop better methods for predicting disease outcome in cancer patients. Specifically, advanced computational algorithms and software developed as part of this project will be used to predict gene regulators and their targets from mRNA and microRNA expression profiles. We propose to study three types of gene regulation in cancer samples: i) transcription of protein coding genes ii) transcription of microRNA genes and iii) translational regulation of protein coding genes by mature microRNAs. The predicted regulatory events will be compared with biological knowledge in pathway databases in order to discover novel regulatory events, detect biological processes characteristic of specific cancer subtypes and assess the accuracy of the prediction algorithms. Such detailed knowledge of gene regulation can improve the prediction of clinical outcomes and guide the choice of therapy for individual patients. Application of MAGNet Tools to Gene Regulatory Networks in Cancer Principal Investigator: Sander, Chris Project Narrative We propose to analyze molecular genetic data from cancer patients using advanced computational methods. The research supports the long-term goal of building genomic profiling tools for predicting disease outcome and response to therapy in individual cancer patients. [unreadable] [unreadable] [unreadable]